Has anyone seen this front-end before? - Page 12 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 21st April 2012, 03:12 PM   #111
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Default suitable name

Quote:
Originally Posted by miralin View Post
Hi Edmond
Congratulate you with inventing and introducing relatively simple circuit with superb linearity.
Hi Mir,

Thank you!

Quote:
I think it can become a standard for the years.
And perhaps even the end of the ubiquitous (and boring!) blameless topology.

Quote:
But why do you call it “transimpedance”? Common base stage is not transimpedance, even if you use a super pair instead of a single transistor, and the whole circuit is definitely transconductance (98 mA/V).
Regarding Q7...Q12 in isolation, you are perfectly right. It is a CCCS with a (current) gain of just 1.
On the other hand, since the output of this stage is tied to a (well defined) impedance, the final output signal takes the form of a voltage source (well, at least, that's the intention). This is even more true for the 2nd figure (the emitters of Q19/20). So, in this respect, we might call it a TIS.
Please, look here for a further discussion.

Quote:
IMHO, it can be called something like “an LTP loaded with current mirror fully cascoded with super pair in a complementary configuration”, where “fully cascoded” means your combination of conventional and folded cascoding, that I’ve never seen before.
This long name is inconvenient, but shows the way to define what each component of the whole circuit gives to excellent performance.
1.We can exclude “complementary configuration” by connecting bases of Q1 and Q6 to the ground (I mean Fig.1 on http://home.tiscali.nl/data.odyssey/Super_TIS.html).
2.We can exclude “fully cascoded” by connecting collectors of Q5 and Q6 to +50v and -50v.
3.We can exclude “super pair” by connecting bases of Q10 and Q11 to V4 and V5 (1.5v).
4.We can make combinations of 1,2,3, each time measuring THD20k.
Would you like to do this?
Sure, but not now. Perhaps next week. But first I have to update my website with a comparison between OMC vs TPC vs TMC.

Quote:
With best wishes – Mir

PS. A short name may be “Stewart Audio Front End” (SAFE) – like “Baxandall Super Pair”. S may also mean “super”.
Well, that's too much honor. Besides, we mustn't forget the input from:
Dimitri Danyuk, "On the Optimization of Enhanced Cascode", AES paper, October 2008.

So, 'BDS front end'?

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 24th April 2012, 10:48 PM   #112
Steven is offline Steven  Netherlands
diyAudio Member
 
Steven's Avatar
 
Join Date: Oct 2003
Location: The Netherlands
Default Baxandall super pair

Hi Edmond,

Nice circuit!
I think I have seen the Baxandall super pair only be used in commercial amplifiers from Yamaha, such as the MX2000, the XP1000/2500/3500, the XS250/350...

Steven
Attached Images
File Type: jpg Yamaha MX2000.jpg (50.2 KB, 317 views)
File Type: jpg Yamaha XP1000.jpg (29.6 KB, 318 views)
__________________
The Analog Art shows no sign of yielding to the Dodo's fate. The emergence and maturation of monolithic processing finesse has perhaps lagged a bit behind the growth of the Binary Business. But whereas digital precision is forever bounded by bits, there is no limit excepting Universal Hiss to the ultimate accuracy and functional variety of simple analog circuits. - Barry Gilbert, 1973
  Reply With Quote
Old 25th April 2012, 02:39 PM   #113
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally Posted by Steven View Post
Hi Edmond,

Nice circuit!
Thank you Steven.


Quote:
I think I have seen the Baxandall super pair only be used in commercial amplifiers from Yamaha, such as the MX2000, the XP1000/2500/3500, the XS250/350...

Steven
You're right, these amps do contain Baxandall super pairs. However, in this case they are used differently: as a TIS (or a VAS or CCCS if you like) which also provide current gain. In case of the "Super TIS", the Baxandall super pairs are use as cascode, which provides no current gain (just 1x). Here, it's the next stage (pre-driver) that provides the current gain.

BTW, I get the impression that among the established audio guru's nobody realizes/recognizes the far reaching implications of moving the gain to the next stage.

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 25th April 2012, 03:28 PM   #114
Waly is offline Waly  United Kingdom
diyAudio Member
 
Waly's Avatar
 
Join Date: Jan 2011
Location: London
Quote:
Originally Posted by Edmond Stuart View Post
BTW, I get the impression that among the established audio guru's nobody realizes/recognizes the far reaching implications of moving the gain to the next stage.
Probably not gonna happen until people are gonna bite the bullet and try to build this thing.

BTW, I've loaded the full TIS plus the output stage (per your web site "Super TIS integrated with an Auto Bias-II Output Stage") in the simulator and had troubles with the stability, for the original TMC compensation values. The global loop is though stable, so it must be some sort of local (Miller?) stability issue (for the moment, I don't have the time to get more in depth). 3x larger TMC caps were required to stabilize the thing under all circumstances - but then the THD20 was "only" 20ppm @4ohm, just before clipping (yes, I know, Cadence sucks, models are different, MC rules, I am ignorant, etc...). Still impressive, but I wouldn't hold my breath to see this working properly, outside the simulator...

And speaking about performances, I wonder if the sheer complexity is worth even a x10 improvement over an optimal Blameless implementation. Is it worth the effort (and cost) to improve from 20ppm to, say, 2 ppm?

There are a few good reasons why TMC "shines" in this configuration, compared to TPC. It's though not an apple to apple comparison here, more perhaps later.

Last edited by Waly; 25th April 2012 at 03:31 PM.
  Reply With Quote
Old 25th April 2012, 04:21 PM   #115
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Walter,

There are many loops, not just global and Miller, but also the differential and common mode loops of the AB2 circuit and the loop of the O/P current limiter. All of these loops have to verified. That means a lot of work.
But first, what were the symptoms of the instability?

>Is it worth the effort (and cost) to improve from 20ppm to, say, 2 ppm?
I can't answer that question. The only thing I can say is that, when squeezing out the last ppm's of distortion, you will inevitably be confronted with the law of "diminishing returns".

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 25th April 2012, 04:45 PM   #116
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally Posted by Waly View Post
Probably not gonna happen until people are gonna bite the bullet and try to build this thing.
[...]
Come on. You don't have to build that thing in order to prove that the super TIS solves two known issues at once that plagued a fully complementary front end. These issues are (or better: were):
1. An ill defined quiescent current of the VASes (see the Randy Slone amp)
2. The issue of the so called 'fighting' VASes (caused by unequal gain of top and bottom VAS and/or unequal compensation caps).

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote
Old 25th April 2012, 05:42 PM   #117
Waly is offline Waly  United Kingdom
diyAudio Member
 
Waly's Avatar
 
Join Date: Jan 2011
Location: London
Quote:
Originally Posted by Edmond Stuart View Post
That means a lot of work.
But first, what were the symptoms of the instability?
Stable at zero input. As soon as a small signal (say, 100mV@20KHz) is added to the input, output goes into oscillation @ about 1MHz.

Quote:
Originally Posted by Edmond Stuart View Post
Come on. You don't have to build that thing in order to prove that the super TIS solves two known issues at once that plagued a fully complementary front end. These issues are (or better: were):
1. An ill defined quiescent current of the VASes (see the Randy Slone amp)
2. The issue of the so called 'fighting' VASes (caused by unequal gain of top and bottom VAS and/or unequal compensation caps).
You answered your own question: nobody cares about squeezing the last ppm out of an audio amp. There are other performance criteria, like "fashionable design" ["zero global feedback", SS and tube mixtures, other buzzword based BS]. Build it, then start BSing about "the wonderful sound of TIS", and with lots of luck you may have a chance to promote the concept.

Different story in IC design for ultra low distortion (e.g. required in DSL drivers, 'cause any drop of noise or distortion maps to extra bandwidth). But then in IC designs you don't have to use TIS to squeeze the last drop of performance. Free matching of devices makes implementing non-gm doubling complementary stages (and also avoiding the issues you are mentioning above) very easy. Not to mention the availability of essentially free transistors with Ft in the GHz range, allowing unity loop gains of hundreds of MHz.
  Reply With Quote
Old 25th April 2012, 08:45 PM   #118
Steven is offline Steven  Netherlands
diyAudio Member
 
Steven's Avatar
 
Join Date: Oct 2003
Location: The Netherlands
Quote:
You're right, these amps do contain Baxandall super pairs. However, in this case they are used differently: as a TIS (or a VAS or CCCS if you like) which also provide current gain. In case of the "Super TIS", the Baxandall super pairs are use as cascode, which provides no current gain (just 1x). Here, it's the next stage (pre-driver) that provides the current gain.
Hi Edmond,

I know Yamaha uses them in a completely different (traditional) way. It was just because Baxandall pairs are almost never used in commercial designs and some discussion arose about stability issues, that I wanted to show these applications of the Baxandall pair.
BTW Already in 2004 these stability issues were discussed on this forum, both as TIS (VAS) and as cascode stage: Baxandall Super Pair. But going through that thread I just saw that also Yamaha's use of it was already mentioned.

I've never seen your use of the Baxandall pair before and I like the way you use it together with the current mirror to create a low-error push-pull current subtractor for the collector currents of the IPS. At first glance the NPN/PNP combination also has some similarity with the Lender VAS, a kind of semi-folded cascode.

Cheers, Steven
__________________
The Analog Art shows no sign of yielding to the Dodo's fate. The emergence and maturation of monolithic processing finesse has perhaps lagged a bit behind the growth of the Binary Business. But whereas digital precision is forever bounded by bits, there is no limit excepting Universal Hiss to the ultimate accuracy and functional variety of simple analog circuits. - Barry Gilbert, 1973
  Reply With Quote
Old 25th April 2012, 09:39 PM   #119
diyAudio Member
 
keantoken's Avatar
 
Join Date: Aug 2006
Location: Texas
Blog Entries: 2
Quote:
Originally Posted by Edmond Stuart View Post
I get the impression that among the established audio guru's nobody realizes/recognizes the far reaching implications of moving the gain to the next stage.
I'm guessing this means a heavily buffered and/or error corrected output stage? Better stability and speed?
  Reply With Quote
Old 25th April 2012, 10:56 PM   #120
diyAudio Member
 
Edmond Stuart's Avatar
 
Join Date: Nov 2003
Location: Amsterdam
Quote:
Originally Posted by Steven View Post
Hi Edmond,
[...]
I've never seen your use of the Baxandall pair before and I like the way you use it together with the current mirror to create a low-error push-pull current subtractor for the collector currents of the IPS. At first glance the NPN/PNP combination also has some similarity with the Lender VAS, a kind of semi-folded cascode.

Cheers, Steven
[bold chars. me]

Hi Steven,

> low error
Indeed, that's an important feature of the circuit. It is important, because the subtractor (or IPS) is the only stage in the whole chain that you can't improve (i.e. lower its distortion) by means of global feedback, as half of this stage stays outside the FB loop. In this respect it's the weakest chain. Therefore, the intrinsic distortion should be as low as possible. I know there are other means, like a cross quad, multiple tanh, etc, but they are more complex.

BTW, do you have a link to the Lender VAS (I've never heard of it )

Cheers,
E.
__________________
Een volk dat voor tirannen zwicht, zal meer dan lijf en
goed verliezen dan dooft het licht…(H.M. van Randwijk)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Bogen Front End lpd Tubes / Valves 2 5th March 2008 05:12 AM
NE5534 with a new front end ...anybody? ashok Solid State 9 2nd June 2006 02:30 AM
My New Front End Pics! micb Car Audio 1 26th May 2006 03:26 PM
Noisily front end mantistube Analogue Source 10 16th January 2006 02:57 AM
Front end for sub amp mashaffer Solid State 9 22nd July 2005 04:35 AM


New To Site? Need Help?

All times are GMT. The time now is 12:45 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2